Kat Gehrke, 25, had no idea that delivering her first baby would lead to the biggest nightmare of her life. On July 5, 2006, Gehrke had a cesarean section at Indian River Memorial Hospital in Vero Beach, Fla.—after more than 20 hours of labor her cervix had not dilated more than 1.6 inches (four centimeters). The procedure seemed to go off without a hitch, and her doctor sent her and new daughter Kaylie home after just two days in the hospital.

All seemed fine, except "my temperature never went back to normal after surgery," Gehrke says. During her first few days at home, she had a low-grade fever that hovered around 100 degrees Fahrenheit (38 degrees Celsius) and she noticed a lump had formed below her incision. By the fourth day, the lump had ballooned to the size of a lime, her fever had jumped to 103 degrees F and her incision was intensely painful. "It was like someone had taken a burning match and stuck it inside" the cut skin, Gehrke says. She immediately went to see her doctor, who took out the staples (as is customary a few days after a C-section) and examined the growing bulge under the wound. He dismissed the pain as normal and prescribed antibiotics for what he diagnosed as a breast infection based on a nickel-size lump that he felt in one breast. But he was wrong. "I didn't have a breast infection at all," Gehrke says, noting that the breast lump was merely a clogged milk duct (common in women who nurse).

A few days later, part of the incision burst open, releasing so much blood and fluid that "the entire [bathroom] floor was soaked in blood," Gehrke says. "My grandma took two beach towels to sop it all up."

Gehrke's grandfather rushed her to Indian River Memorial's emergency room where doctors and nurses cleaned the wound and packed it with gauze to allow for drainage. They sent a sample of the fluid to the hospital lab for analysis and called Gehrke's ob/gyn to inform him that the infection was probably caused by the bacterium Escherichia coli. Gehrke's ob/gyn prescribed sulfa antibiotics, which are commonly used to treat E. coli infections. Two days later, the lab results came back, revealing that Gehrke did not have E. coli, but rather a staphylococcus, or staph, infection caused by methicillin-resistant Staphylococcus aureus (MRSA), a potent bacterium that has developed resistance to most of the old standby antibiotics, making it difficult to treat and potentially fatal.

MRSA causes some 94,000 invasive infections in the U.S. each year, resulting in almost 19,000 deaths—more than those caused by human immunodeficiency virus (HIV)—said a study published this week in JAMA The Journal of the American Medical Association. And "the majority of these cases appeared to be health care–acquired," says Elizabeth Bancroft, a medical epidemiologist with the Los Angeles County Department of Public Health and author of an editorial that accompanied the study.

Learning the infection was MRSA, Gehrke's gynecologist immediately switched her to a stronger antibiotic and put her on bed rest; nurses from the Visiting Nurses Association came to her house daily to pack the wound in gauze and check her vital signs. But the swelling remained and the wound continued to ooze pus.

After three months of this, the wound still had not healed. At the advice of her ob/gyn's partner (who was filling in for her doctor that day), Gehrke went to see doctors at Indian River Hospital's wound care facility. They told her she needed a second operation to remove the tissue destroyed by the infection. Surgeons reopened her incision and discovered a festering infection that had caused extensive damage. It was "like looking at a hole in your belly [that is] seven inches wide and six-and-a-half inches deep," she recalled in an interview with ScientificAmerican.com. After the operation, Gehrke stopped seeing her ob/gyn, but continued to be treated by the wound care physicians and visiting nurses. She says she was mostly bedridden for another three months because it was painful to move while attached to a wound V.A.C., a suction device that aids healing by vacuuming pus, blood and other fluids.

Gehrke survived but it took seven months when all was said and done for the infection to clear up and the wound to heal. She says she was bedridden for a total of six months and racked up $13,000 in out-of-pocket expenses for home care and procedures associated with her infection.

"My credit is shot," says Gehrke, who works as a server at a local restaurant and whose husband works as a diesel mechanic for Wal-Mart. Their combined salaries amount to about $34,000 annually before taxes.

Gehrke asked her ob/gyn why this had happened. "'These things happen in hospitals' is pretty much what he told me," she says.

It is difficult to pinpoint the source of Gehrke's infection. It may have stemmed from a dirty instrument used during her C-section or from unwashed hands or the contaminated gloves of a health care worker. But one thing is almost certain: she picked up the bug at the hospital.

Gehrke is one of millions of patients who have unwittingly contracted infections in hospitals, where they went expecting to get well—but instead got sicker. Every year nearly 100,000 people die of infections they developed in U.S. hospitals and healthcare facilities, a greater number than those killed in homicides and car accidents combined. Some 1.7 million patients contract hospital infections annually, according to the most recent data from the Centers for Disease Control and Prevention (CDC).

Many of these infections are caused by multidrug-resistant superbugs such as MRSA and vancomycin-resistant enterococci (VRE). Heavy use of antibiotics in hospitals encourages the emergence of stronger and stronger bacteria. Exposing a bacterial strain to one antibiotic essentially weeds out the weak and selects the hearty bugs that can survive. Then the next generation of antibiotics is called on; eventually the bugs become resistant to that as well and the bacteria continue evolving until eventually no antibiotic can kill them. "You can end up with bugs that we really don't have medications to kill," says Allison Aiello, assistant professor of epidemiology at the University of Michigan School of Public Health. Experts estimate that more than 70 percent of all hospital-acquired infections are caused by bacteria that are resistant to at least one of the drugs commonly used to treat them.

Hospitals not only provide optimum conditions for the evolution of superbugs, but they also provide a plethora of inviting pathways for bacteria to get inside human bodies: open wounds from surgical incisions, catheter tubes running in and out of blood vessels and urinary tracts, and ventilators inserted through noses or throats and into windpipes.

What's most shocking about hospital infections, experts say, is that most of them can be avoided. "The vast majority of all hospital infections are preventable," Bancroft says. In the past, "the mantra was that hospital infections are inevitable," she says, but the attitude is changing because many hospitals have proved it wrong.

In 1978 the University of Virginia (U.V.A.) Medical Center in Charlottesville had its first case of MRSA. The bug spread from patient to patient despite the fact that health care workers were washing their hands after touching bodily fluids as well as donning gowns, gloves and masks when caring for patients with clear signs of MRSA infections (such as pus-discharging wounds or pneumonic coughing), says Barry Farr, who was a medical resident at the time and is now professor emeritus of U.V.A.'s Department of Medicine. By 1980, nearly half of the hospital's staph infections were caused by MRSA.

In an effort to control the problem, the hospital decided to actively seek out and isolate not only patients infected with MRSA but also those who were colonized, meaning they carried the bug on their skin or inside their noses, sputum or urine. (People who are colonized may be carrying millions, if not billions, of bacteria that can easily spread to others, either through direct contact or by touching common surfaces such as bed rails, doorknobs and blood pressure cuffs). U.V.A. began testing all high-risk patients for MRSA infection and colonization; those who tested positive were placed in contact-isolation areas with warning signs on their doors alerting health care workers of the patients' contagious status and instructing them to wash their hands after touching them. (The CDC did not even recommend hand washing before and after all patient contacts at this time.) Using active detection and isolation, U.V.A. had totally wiped out MRSA within 18 months, Farr says. "I watched this work at U.V.A. in 1980 to 1982" and "there was no question that it worked."

Despite ongoing efforts by U.V.A. and a handful of other U.S. hospitals to identify and isolate colonized patients, MRSA raced through the health care system virtually unchecked because most facilities lacked effective infection-control programs. By the 1990s, "the whole health care system became completely permeated by [MRSA]," says Farr, who for the past three decades has been a leading proponent of active detection and isolation to control superbugs. In 1995 MRSA infections accounted for 22 percent of all health care–associated staph infections, compared with only 2 percent in 1974.

Today, "close to 70 percent of staphylococcus causing infections in intensive care units are MRSA," says Fred Tenover, acting director of the CDC's National Center for Infectious Diseases Office of Antimicrobial Resistance, noting that within that average there is huge variation from one hospital to the next.

The University of Virginia is one of many exemplars of infection control in the U.S. Between 2001 to 2005, a group of 32 Pennsylvania hospitals working with the CDC slashed the rate of central lineassociated bloodstream infections (associated with catheters placed in veins) in intensive care units by 68 percent. One of the participants in that initiative, Allegheny General Hospital in Pittsburgh, reduced the total number of these infections by nearly 90 percent in one year (2003 to 2004) and the unit has recently gone nearly 18 months without an infection. The dramatic decline at Allegheny occurred despite a near doubling in the use of catheters and a steady increase in the severity of illness of patients in its intensive care unit. "Using more catheters and caring for sicker patients are not justifications for higher numbers of infections," Richard Shannon, then chair of Allegheny's department of internal medicine, told a congressional panel last year that was considering proposals designed to reduce hospital-acquired infections.

Shannon, who has since become chair of the department of medicine at the University of Pennsylvania in Philadelphia, outlined his successful campaign during a recent interview with Scientific American.com. "We took the Toyota production system and applied it to placement and maintenance of catheter insertions," and later ventilator insertions, he says. To create a perfect product, "all things have to be executed perfectly," and everyone in the organization must be on board, he says.

The same is true for infection control. In order to determine the perfect way to execute each step of placing and maintaining catheters, Shannon says he tapped the doctors, nurses and medical technicians in his hospital for advice. After developing standardized procedures for everything from putting on a hospital gown and cap to washing hands to actually placing the catheter, "we trained people and observed them" instead of just handing out a document with instructions, he notes, adding that leadership is a key component of eradicating the problem. "This will never occur," he says, "until a senior person says, 'I've had enough.'"

Despite the success of these hospitals and several others, critics say the U.S. lags far behind some countries in keeping superbugs in check. Denmark's MRSA prevalence reached 33 percent in the 1960s but steadily declined after a strict infection-control policy was implemented; it has remained below 1 percent for 25 years. In the Netherlands and parts of Scandinavia health care–associated MRSA prevalence has been maintained at 1 to 3 percent for decades through an aggressive "search and destroy" approach: Hygiene practices are strictly enforced and all high-risk patients and staff members are systematically screened for MRSA; those found to be infected or carriers are quarantined.

Many experts believe that the search-and-destroy method is the best answer to America's superbug problem. "We need to start looking for these [drug-resistant strains] on admission and putting patients in isolation," says Marcia Patrick, spokesperson for the Washington D.C–based Association for Professionals in Infection Control and Epidemiology (APIC).

Most hospitals follow CDC guidelines, which have not explicitly recommended routine active surveillance cultures to identify and isolate patients with superbugs, even though more than 140 studies have shown that MRSA and VRE can be controlled this way, Farr says.

Interestingly, the CDC advises routine HIV screening of adults, adolescents and pregnant women in health care settings, but does not recommend universal testing for MRSA, which kills more people than HIV.

Meanwhile, physicians say that it is tough to get health care workers to take even the simplest precautions such as scrubbing their hands between patients. Patients will no doubt be stunned to learn surveys show that compliance with such rules now hovers at only around 50 percent on average at medical institutions in this country.

"We have the knowledge to prevent these, and what has been lacking is the will, the energy, the pressure to do it," says former lieutenant governor of New York, Betsy McCaughey, chair of the nonprofit Committee to Reduce Infection Deaths (RID). But the climate is changing fast, she adds.

"Hospitals are moving into a new legal environment," she says. "As long as it was believed that hospital infection was the inevitable risk you faced in the hospital and nothing could be done about it, hospitals were protected from lawsuits." But that is no longer the case as more and more people come to recognize that these infections are mostly preventable.

The law itself has begun to change. U.S. Rep. Tim Murphy (R–Pa.) has proposed the Healthy Hospitals Act of 2007, which would require hospitals to publicly report their infection rates as well as provide economic incentives to those that successfully reduce their rates. Since 2003 19 states have passed legislation requiring hospitals to report their infection rates: Colorado, Connecticut, Delaware, Florida, Illinois, Maryland, Minnesota, Missouri, New Hampshire, New York, Ohio, Oregon, Pennsylvania, South Carolina, Tennessee, Texas, Virginia, Vermont and Washington. These laws vary dramatically: some states impose penalties as high as $1,000 per day for hospitals that fail to comply, whereas others have almost no enforcement power, says Lisa McGiffert, manager of Consumer Union's StopHospitalInfections.Org, a project aimed at getting states to publish infection rates of hospitals and other health care facilities. (The site provides recent information on hospital infections and an online forum where over 1,800 victims have shared their experiences. It also encourages citizens to sign petitions and write letters to hospitals and government officials.) Reporting infection rates "stimulates activity," McGiffert says. "Once a bill passes in a state, hospitals start getting better," she says, noting that it's still too early to see a cause-and-effect relationship.

But Michael Bennett, president of the lobbying group Coalition for Patients' Rights (CPR), says that "hospitals have a huge disincentive to accurately report their infections," and "there are countless ways that a reporting system can be gamed." Reporting laws are good, Bennett says, but "there is no data that I'm aware of that suggests reporting has lowered infection rates or the numbers of infections."

RID's McCaughey says that hospitals need to take the problems seriously and not skimp when it comes to forking out funds to do the job properly. In addition to saving lives, she notes that "We can show hospitals that they can be more profitable by preventing infections." She estimates that hospital infections add at least $30.5 billion a year to the nation's health care tab in hospital costs alone—enough to pay for the entire Medicare (Part D) prescription drug program. According to a recent article in The Lancet, "virtually all published analyses that have compared the cost of screening of patients on admission and using contact precautions with colonized patients" show that "the costs of caring for patients who become infected with MRSA are much greater than the costs of screening programs."

What this country needs is federal infection-control legislation derived from evidence-based best practices, Bennett says. He co-founded CPR three years ago after his father died at age 88 from a deadly infection contracted at Northwest Hospital Center and Sinai Hospital, both in the Baltimore area, while being treated for a nonlife threatening respiratory infection. Bennett is putting together a task force of scientists to draft new legislation or add teeth to existing measures that would mandate infection-control guidelines. That language would surely include active detection and isolation for the most egregiously out-of-control superbugs like MRSA. Hospitals need to "screen high-risk admissions and treat them proactively," Bennett says. "We are talking about life and death and immense human suffering that has come about through systemic negligence."